update: preemptive kernel for O(1) sched
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 3 lines · Dieter Nützel
what about lock-break?
I am running your former one as always with
lock-break-rml-2.4.18-pre1-1.patch ...;-)
I haven't tested O(1) together with lock-break personally, yet, but I have a confirmation of success from a couple of users. There are no reasons it shouldn't work.
Quoted 1 line
Any success together with AA vm-22?
Again I don't have any personal tests. I do know, however, the Rik's rmap VM works great.
I suspect Andrea's updated VM from -aa will work, but there may be other parts of his patch (RCU) that are not preempt friendly.
Quoted 1 line
What about latencytest0.42-png for latency testing?
A good test.
Robert Love
Quoted 8 lines · you
Quoted 3 lines · Dieter Nützel
what about lock-break?
I am running your former one as always with
lock-break-rml-2.4.18-pre1-1.patch ...;-)I haven't tested O(1) together with lock-break personally, yet, but I have a confirmation of success from a couple of users. There are no reasons it shouldn't work.
I'll say that I am running this altogether.
Only vm-22 or rmap-11 are missing for now.
2.4.17 was running for me with vm-21 and preempt+look-beak.
Quoted 4 lines
Quoted 1 line
Any success together with AA vm-22?
Again I don't have any personal tests. I do know, however, the Rik's rmap VM works great.
AA, too . Best throughput I've ever seen.
Quoted 2 lines
I suspect Andrea's updated VM from -aa will work, but there may be other parts of his patch (RCU) that are not preempt friendly.
But I've only used the VM parts not RCU, etc.
Quoted 3 lines
Quoted 1 line
What about latencytest0.42-png for latency testing?
A good test.
Expect some numbers (latency) for O(1) with and without preempt+lock-break maybe with AA-VM, soon.
-Dieter
Quoted 3 lines · Dieter N?tzel
Quoted 3 lines
what about lock-break?
I am running your former one as always with
lock-break-rml-2.4.18-pre1-1.patch ...;-)
Quoted 3 lines · Robert Love
I haven't tested O(1) together with lock-break personally, yet, but I have a confirmation of success from a couple of users. There are no reasons it shouldn't work.
I have at least run it on my laptop, together with rmap even. No pathological behavior that I can tell. Of course, the interactive response is wonderful, but I haven't precisely measured anything, as I have enough other things to measure precisely it's a bit far afield.
Cheers,
Bill
preempt-kernel for 2.5.2:
ftp://ftp.kernel.org/pub/linux/kernel/people/rml/preempt-kernel/v2.5
Note, again, this patch is not compatible with later O(1) patches due to the load_balance changes. I'll make the changes as they are merged into 2.5 proper.
No other changes in this release.
Robert Love